Research map / 3D vision / Face and shape
Synchronous Diffusion for Unsupervised Smooth Non-Rigid 3D Shape Matching
Dongliang Cao, Zorah Laehner, Florian Bernard
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add synchronous diffusion regularisation L_diff (ablation) | unsupervised deep functional maps · TOPKIDS | geodesic error: 9.2 -> 5.4 | Improved |
| randomly sample diffusion times (multiscale) instead of a fixed time (ablation) | L_diff · TOPKIDS | geodesic error: 7.8 -> 5.4 | Improved |
| random initial distributions instead of Laplacian eigenfunctions (ablation) | L_diff · TOPKIDS | geodesic error: 10.7 -> 5.4 | Improved |
| diffusion regularisation instead of heat kernel matching (ablation) | L_diff · TOPKIDS | geodesic error: 8.9 -> 5.4 | Improved |
| diffusion regularisation instead of Dirichlet energy (ablation) | L_diff · TOPKIDS | geodesic error: 9.5 -> 5.4 | Improved |
| diffusion regularisation instead of cycle-consistency loss (ablation) | L_diff · TOPKIDS | geodesic error: 9.8 -> 5.4 | Improved |
| increase maximum diffusion time from 1.0^-2 to 1.0 (ablation) | L_diff · TOPKIDS | geodesic error (x100): 5.4 -> 23.7 | Hurt |
| decrease maximum diffusion time from 1.0^-2 to 1.0^-4 (ablation) | L_diff · TOPKIDS | geodesic error (x100): 5.4 -> 10.3 | Hurt |
Closest papers by meaning
- Fast Markov Random Field Optimisation for Topologically Noisy 3D Shape Matching (CVPR 2026 · similarity 0.89)
- Denoising Functional Maps: Diffusion Models for Shape Correspondence (CVPR 2025 · similarity 0.88)
- DiffuMatch: Category-Agnostic Spectral Diffusion Priors for Robust Non-rigid Shape Matching (ICCV 2025 · similarity 0.88)
- Probabilistic Deformation Consistency for Unsupervised Shape Matching (AAAI 2026 · similarity 0.88)
- Integrating Efficient Optimal Transport and Functional Maps For Unsupervised Shape Correspondence Learning (CVPR 2024 · similarity 0.86)
- SpiderMatch: 3D Shape Matching with Global Optimality and Geometric Consistency (CVPR 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
